Some test results using shadow images with mask

IsHYuhi / ST-CGAN_Stacked_Conditional_Generative_Adversarial_Networks

Unofficial implementation of ''Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removal'' with PyTorch

67 stars 15 forks source link

Some test results using shadow images with mask #13

Closed mychina75 closed 3 years ago

mychina75 commented 4 years ago

Hi, thank you for your work. I tested pretrained G2 model with some shadow image, together with masks. here are some results, any idea of getting improvements?

thank you.

grid_SBU-test34 grid_SBU-test33 grid_SBU-test29 grid_SBU-test22

IsHYuhi commented 3 years ago

I apologize for the long delay in replying. The module is not implemented in this code, but you need to observe the metrics or loss for validation data and save the model. Please reopen this issue if you have any problems.

Thank you.